Nonparametric algorithms reduce "the need to choose features very carefully" (I guess that makes sense if you think of features as mathematical transformations on stuff observed rather than stuff observed in general... a nonparemetric algorithm surely can't avoid the fact that you left something off, though I guess it can help avoid the fact that you threw a bunch of extra stuff in...)Continue reading →
n.b. the math is all in code blocks because the markdown processor screws with underscores and carets otherwise, and mathjax can't handle that. This is making me insane and I might actually write some kind of post-processor to jerk around the generated html files to fix this, but it'll have to do for now.Continue reading →
I know a lot more about game theory than I do about graphs/trees/etc. (Political scientist, yo.) So here's an attempt at a translation of the idea of backwards induction into tree terms.
First, intuition/background: backwards induction is a method for finding subgame perfect equilibria of a sequential game in extensive form. (For simplicity, let's assume a perfect information game.) The short version is that, starting at the terminal node, you set the last player's choice for that node as the choice that yields the best they can achieve given the history of play leading up to that node. With that information, you now know the values for each choice of the second-to-last node for the second-to-last player, so set their choice to that. And so on, inductively.Continue reading →
Here's a trick I just cooked up (well, it's probably widely known by, like, everyone on earth who regularly does css stuff, but I just discovered it for me, so I'll take a tiny bit of credit.).
Suppose you want to start off your site with an icon screen, like an iPhone or something. (I'm actually basing the design of my new website, in progress, on the old Palm Pilot home screen. Though I just moved away from the green, so maybe not anymore.)Continue reading →